Party Affiliation, District-Level Incentives and the Use of Parliamentary Questions in Chile's Presidential Democracy

Abstract

The issuance of parliamentary questions (PQ) in presidential democracies reflects an effort to connect with the electoral constituency to advance the legislator's career. We postulate six hypotheses on the association between party affiliation, career advancement and district-level incentives and the issuance of PQs in Chile's presidential multiparty democracy. We test them using a novel dataset containing 68,424 inquiries (oficios legislativos) issued by Chamber of Deputies legislators in three legislative terms (2006–2018). Though district-level variables play a role in the issuance of PQs, incentives of political ambition do not. As opposition coalition legislators make more use of PQs than ruling coalition legislators, there is preliminary evidence to associate PQs with a possible oversight role.

As a widely used tool in most parliamentary democracies, parliamentary questions (PQs) are non-legislative acts that can be used by the minority to check on the actions of the legislative majority and by all legislators to advance their careers and other goals. PQs also exist in some presidential democracies, where they have been associated with career-advancing goals, such as name recognition and constituency service (Alemán et al. 2018). But because PQs force the executive to disclose information on its actions and decisions, legislators can also potentially use PQs to exercise a constitutionally mandated accountability function.

Between 2006 and 2018, the Chilean executive had strong proactive legislative powers and the institutionalized party system coexisted with electoral rules that promoted the cultivation of a personal vote. We use the issuance of PQs to examine if, in addition to the previously reported constituency service and career-advancement goals, party affiliation is associated with the issuance of PQs – an observation that would lead one to believe that PQs can also potentially be used as a tool to exert accountability on the national government.

To account for party-level incentives, we propose two hypotheses that associate the issuance of PQs with the legislator's affiliation to the ruling coalition and to the president's party. For individual- and district-level incentives, we postulate four hypotheses that associate the issuance of PQs with district features and traits of legislators. After presenting the case of Chile, we explain our methodology, the dataset (see the Supplementary Material) and the inferential analysis. We finish by discussing the implications of our findings.

The uses of parliamentary questions

Since they are more prominent in parliamentary systems (Martin 2011), PQs are generally studied as tools for constituency service and career advancement. Whether they are written or oral, their timing (submitted in advance or spontaneous), the option to join the debate (if other members can speak) and to vote on a motion all impact their use (Russo and Wiberg 2010: 218–219). When legislators can issue written PQs, their parties are more limited in how they can restrict the scope of what can be covered (Rasch 2011; Rozenberg and Martin 2011). In countries where legislators can introduce written and oral PQs, the latter are more important (Rasch 2011).

Sometimes legislators use PQs to obtain information for use in legislative committees (Martin 2011) or future bills (Bailer 2011). Olivier Rozenberg et al. (2011) find that PQs are used to channel political conflict between the government and the opposition in Germany, internal dissent in political parties in the United Kingdom, reward policy expertise and committee specialization in Spain, and to represent local issues in France. Others have underlined the role of PQs as an information-gathering mechanism (Proksch and Slapin 2015), a trigger to alert the audience to possible violations of EU rules by member states (Jensen et al. 2013), a tool used by opposition party dissidents to express their support for the national party positions (Proksch and Slapin 2015) or to exert oversight over agencies that deal with more visible issues (Font and Pérez Durán 2016). Opposition party legislators in the EU are systematically more likely to use PQs.

In presidential systems where the executive enjoys legislative proactive powers – like agenda setting (Cox and Morgenstern 2001) – legislators complement their limited reactive powers by creating investigative committees or issuing PQs (Martin 2011). PQs can also be used by parties to assert issue ownership and foreground their priorities (Otjes and Louwerse 2018). This would explain the differences in the use of PQs by coalition partners, especially for minority members of a government coalition. Previous studies on the use of PQs in presidential systems focus on career-advancing and constituency-connection motives (Alemán et al. 2018; Chasquetti and Micozzi 2014). Here, we look at whether party affiliation impacts the issuance of PQs. If this is the case, then PQs might be used as a tool to hold the presidential cabinet to account (Moreno et al. 2003: 86–87; Pelizzo and Stapenhurst 2006).

Determinants of the issuance of parliamentary questions

In parliamentary systems, there is discrepancy over the impact of party affiliation on the issuance of PQs. In Ireland, government party legislators are more likely to issue PQs (Martin 2011), while in Belgium and the United Kingdom, opposition party legislators or those who are ideologically more distant from the executive are more likely to do so (Cole 1999; Dandoy 2011; Otjes and Louwerse 2018). The discrepancy can be attributed to the content of the oversight toolbox available to legislators. If in addition to allowing legislators to connect with their electoral constituencies, PQs can be used as tools to exert horizontal accountability (O'Donnell 1998), we would expect ruling coalition legislators to make less use of PQs than members of the opposition. Consequently, our first hypothesis postulates that:

Hypothesis 1: Legislators from opposition parties issue more PQs than legislators from the ruling coalition.

In multiparty coalition governments, parties use their junior ministers to check on the actions of ministers from other parties who occupy more important cabinet posts (Lipsmeyer and Pierce 2011; Thies 2001), when there is a bigger ideological gap between coalition parties (Lipsmeyer and Pierce 2011), and with ministries that deal with issues sensitive to party activists (Greene and Jensen 2016). The check function will also depend on the strength of the committee system (Lipsmeyer and Pierce 2011) and on the distribution of committee presidencies (Kim and Loewenberg 2005). When the committee system is strong and the presidency of the respective committee is not occupied by a member of the minister's party, intra-coalition oversight will be conducted in the legislature and not within the cabinet. Either way, there is strong evidence that multiparty coalition members seek to exercise mutual accountability. Although they are partners in the ruling coalition, parties will compete against each other for marginal seats in the next election.

The presence of broad multiparty coalitions in parliamentary systems, like the European Parliament, hinders the use of PQs (Russo and Wiberg 2010). In presidential systems with multiparty coalitions, however, the dynamics might be different as parties aspire to win more seats as much as they aspire to help their coalitions remain in power. Thus, multiparty presidential democracies experience competition between and within coalitions. As parties can use PQs as a negative campaign tool to expose their direct competitors (Otjes and Louwerse 2018), PQs serve as a mutual control mechanism within the ruling parties. Thus, the motivations of the different ruling coalition parties differ, as legislators from the president's party should be less inclined to issue PQs to the national government than legislators from other ruling coalition parties. This brings us to our second hypothesis:

Hypothesis 2: Legislators from the president's party are less likely to issue PQs than other ruling coalition legislators.

As legislators are motivated by career advancement and constituency service, those incentives also affect their behaviour, independently of their party affiliation (Rozenberg and Martin 2011; Saalfeld 2011). PQs can help build name recognition and establish issue ownership (Martin 2011). In closed-list proportional representation systems, legislators who have a distinct constituency use PQs to cultivate a personal vote (Russo 2011) – which can, presumably, help them secure a better place in their party list in the next election. District-based issues can also explain the issuance of PQs (Blidook and Kerby 2011). In electoral systems that promote the cultivation of a personal vote, legislators have incentives to issue PQs related to their districts' concerns (Alemán et al. 2018; Chiru 2018; Russo 2011). As legislators with longer tenure normally have more name recognition and have developed stronger and more diversified constituency networks, we would expect legislators with less experience to make more use of PQs. Consequently, we postulate that:

Hypothesis 3: First-term legislators send more PQs than legislators with longer tenures.

Political ambition matters in explaining the issuance of PQs. Politicians can have progressive (higher office), static (same office) or discrete ambition (retirement) (Schlesinger 1966). More ambitious and less experienced Swiss legislators are more likely to issue PQs (Bailer 2011). Incumbents who issue a larger number of constituency service-related PQs benefit electorally (Chiru 2018). Legislators with progressive ambition adjust their behaviour to cater to the constituencies they aspire to represent (Hibbing 1986; Van Der Slik and Pernacciaro 1979). In Chile, Chamber of Deputies legislators who aspire to run for the Senate issue more PQs related to the regions they hope to represent (Alemán et al. 2018). Thus, our fourth hypothesis suggests that:

Hypothesis 4: Legislators with static and progressive ambition send more PQs than those with discrete ambition.

Legislators who are more concerned with their re-election are also more likely to try to build name recognition and provide constituency service. Legislators are more likely to ask district-related questions when they feel electorally threatened and when the issue can be articulated so as to resonate in the district (Blidook and Kerby 2011). UK House of Commons members who won by a slimmer margin are also more likely to present more PQs (Kellermann 2016) and so are legislators in Chile's presidential system (Alemán et al. 2018). Thus, our fifth hypothesis suggests that:

Hypothesis 5: Legislators with a lower vote share in the past election send more PQs.

The centre–periphery cleavage also affects the issuance of PQs, with legislators from districts further away from the capital using them more than legislators from central districts (Martin 2011). In supranational units like the European Parliament, legislators from peripheral countries also make more use of PQs (Raunio 1996). In Chile, legislators from faraway districts issue more PQs (Alemán et al. 2018). Local issues impact the issuance of PQs in parliamentary and semi-presidential systems (Rozenberg et al. 2011). When the electoral system promotes the cultivation of a personal vote, legislators from distant districts are more likely to obtain name recognition by using this tool. Thus, we propose a sixth hypothesis that states:

Hypothesis 6: Legislators from districts far away from the capital city send more PQs.

The place of PQs in Chile's legislative process

Since the return of democracy in 1990 and at least until 2018, Chile has had a strong presidential system with stable multiparty coalitions. The president has broad proactive legislative powers – including bill initiation and agenda setting. In turn, the highly institutionalized legislature has mostly reactive powers (Cox and Morgenstern 2001), as it can effectively block or delay presidential initiatives. Presidents who enjoy majority support in the legislature are more successful in advancing their agendas (Alemán and Navia 2009).

Throughout the period, Chile had an open-list proportional representation electoral system with an across-the-board magnitude of 2. In most districts, the two dominant coalitions received one seat each. Thus, elections were mostly an intra-coalition competition. Under proportional representation with closed lists, there are fewer incentives for the cultivation of a personal vote, as party leaders can exert more discipline over their legislators (Proksch and Slapin 2015). In turn, in single-member districts, party leaders are in a weaker position to influence the legislators' activities. Open-list proportional representation systems lie between the two extremes, with limited control capacity for party leaders (Proksch and Slapin 2015), but when the district magnitude is small, incentives to cultivate a personal vote are stronger (Carey and Shugart 1995; Crisp et al. 2004). This would give legislators more discretion on how they can use PQs.

From 1989 to 2017, two stable multiparty coalitions controlled most seats in the legislature, the centre-left Concertación and the centre-right Alianza. The Concertación, comprising the Christian Democratic (PDC), Radical (PR), For Democracy (PPD) and Socialist (PS) parties, controlled the presidency from 1990 to 2010. In 2014, with the addition of the Communist Party (PC), the coalition renamed itself New Majority (NM) and won the presidency for another term (2014–2018). The Alianza, comprising National Renewal (RN) and the Independent Democratic Union (UDI), controlled the presidency for one term (2010–2014).

Coalitions are cohesive in their roll-call votes in Congress, but the government coalition is more cohesive than opposition parties (Toro Maureira 2007). Legislators use bill sponsorship to advance their political careers (Escobedo Aránguiz and Navia 2020) and build cooperation networks across party lines (Gamboa and Toro 2018). Legislators display behaviour that points to constituency service and to the cultivation of a personal vote in combination with party loyalty concerns (Dockendorff 2019, 2020; Gamboa and Toro 2018).

However, since the interests of the parties and their coalitions sometimes differ, intra-coalition tensions arise. Ruling coalition legislators from minority parties might want to distance themselves from unpopular presidents or unpopular government issues. On those occasions, legislators can issue PQs to signal their differences from the government.

The Chilean Chamber of Deputies has a constitutional oversight mandate – in constitutional Article 52 – on the actions of the executive. The Chamber can create investigative committees, summon cabinet ministers to respond to questions, and issue requests for information – PQs – that the government must respond to within 30 days. The reach of the constitutional PQ power is fully detailed in Article 9 of the Organic Law of the National Congress and regulated by Article 309 of the Chamber of Deputies Rules and Procedures. Oversight PQs (oficios de fiscalización) can be directed to any public entity – not only the national government (Article 310, Rules and Procedures). A legislator can announce a PQ from the floor of the chamber during the hora de incidentes (daily logs). Since there are few legislators present during that time, the one-third approval threshold for PQs to be issued is normally easily met.

Deputies can also issue information PQs (oficios de información). According to articles 308 and 309 of the Rules and Procedures, those PQs can be issued directly from the chamber floor, through a permanent committee or via the secretary of the Chamber, and do not require the approval of the Chamber. Information PQs can be directed to any entity – public or private – or any person. Information PQs do not mandate the recipient to respond, but legislators can use the floor of the chamber as a platform to bully them.

Since oversight PQs are normally approved by the Chamber as a package without objections, in practice legislators can use their discretion to issue either information PQs or oversight PQs. The Chamber records do not always identify which type of PQ was issued, though in many cases it can be inferred from the PQ's content. Legislators have increasingly issued PQs to non-national government entities, including autonomous state agencies, local governments, public universities and even the private sector. However, the destination of most PQs remains the national government. We return to the destination of PQs below.

Unlike many parliamentary systems, where legislators use PQs to inquire about the government's legislative priorities, in Chile's presidential system, PQs are not used to address legislative matters. That is regularly done at the committee level or by introducing non-binding resolutions (proyectos de acuerdo) asking the government to move forward on a stalled bill or to introduce a bill on issues where the executive has exclusive bill-initiation powers.

The literature on parliamentary systems points to oral PQs serving more for publicity purposes and written PQs used for more detailed policy content (Rozenberg and Martin 2011: 395). In Chile, as oral PQs are delivered in the Chamber of Deputies at a time when other legislators and the media are mostly absent, the impact of oral and written PQs is similar. In fact, all PQs are eventually delivered as written documents. Additionally, as the number of PQs has increased, their marginal impact has declined. As the public impact of a PQ depends on the media disseminating it to the public, legislators care more about promoting their PQs to the media than delivering them orally in the legislature.

Anecdotal evidence shows that PQs are used for constituency service. On 4 April 2017, right-wing legislator Joaquín Lavín issued a written PQ to the undersecretary of crime prevention inquiring about crime-fighting measures near a supermarket in Maipú, the largest municipality in his district. In his first term, Lavín issued 207 PQs – only two using hora de incidentes. On 16 June 2010, opposition party legislator Marcos Espinoza (PR) issued a PQ to the Ministry of Health inquiring about the construction of a public hospital in Calama, a city in his district. A previous study on PQs in Chile points to their constituency service use, especially by legislators with progressive ambition, from peripheral districts and those who are more electorally vulnerable (Alemán et al. 2018). That work reports that being an opposition party legislator has no effect on the issuance of PQs.

There is anecdotal evidence that shows that PQs are also used as an oversight tool, even by ruling coalition legislators, and that their scope responds to national concerns. For example, on 5 May 2016, Deputy René Saffirio, from the PDC, a ruling coalition minority party, used the hora de incidentes to issue a PQ to the National Children's Service (SENAME) inquiring about deaths of minors under the supervision of that national government agency. Similarly, on 16 March 2017, opposition coalition legislator José Antonio Kast, a member of the UDI and a stalwart conservative, issued – without using the hora de incidentes – a PQ asking for information on sex education policies from the Ministry of Education. During that term, Kast issued 1,081 PQs, far more than the 5 and 56, respectively, he had issued in the previous two terms. In late 2017, Kast ran as an independent conservative presidential candidate on a family-values platform.

Methodology

The unit of analysis is the member of the 120-member Chamber of Deputies by four-year terms between 2006 and 2018. As occasional vacancies are replaced, there were 123, 123 and 121 legislators in each term, respectively. Table 1 shows the composition of the Chamber of Deputies by party in each term. We have 367 legislators for the three terms (but only 210 different persons).

Table 1. Party Composition in the Chamber of Deputies of Chile, 2006–2018

Source: The authors, with data from Chamber of Deputies.

Notes: Conc = Concertación; NM = New Majority; PC = Communist Party; PDC = Christian Democratic Party; PPD = Party For Democracy; PR = Radical Party; PS = Socialist Party; RN = National Renewal; UDI = Independent Democratic Union.

Using 4,475 parliamentary questions from the 2006–2010 legislative term obtained from the hora de incidentes of the Chamber of Deputies in Chile, Eduardo Alemán et al. (2018) found that PQs serve as tools for legislators to connect with their electorate. Andrès Dockendorff (2019), using 2,370 parliamentary questions from hora de incidentes in the 2006–2010 term, reports, discussing it as a control variable, that opposition legislators are more likely to issue PQs.

We compiled a much larger dataset (see the Supplementary Material for this article) based on all PQs issued in every term – not just those in the hora de incidentes. We obtained the PQs using an active transparency initiative request. We received the data for the 2006–2014 period in December of 2015 and for the 2014–2018 period in June 2020. The data received do not distinguish between information PQs and oversight PQs, but include the destination of the PQs – which allows us to distinguish between PQs issued to the national government, national agencies and other state and non-state entities.

The dataset is made up of 11,234 PQs for the 2006–2010 term, 14,499 for the 2010–2014 term and 42,691 for the 2014–2018 term, giving a total of 68,424. There was a sharp increase in the number PQs in 2014–2018. We have almost three times as many PQs for the 2006–2010 term as those retrieved by Alemán et al. (2018) and four times as many as those analysed by Dockendorff (2020). As we have data for several terms, we can compare across terms under different governments and analyse the issuance of PQs based on whether a legislator was a member of the ruling coalition or the opposition. Although rules regulating the issuing of PQs remained unchanged for the entire period, on average, legislators issued four times more PQs in the 2014–2018 term than in 2006–2010. Our dataset includes all PQs – not just those mentioned in hora de incidentes in the floor of the chamber.

We use two indicators for the dependent variable. First, the number of PQs issued by each legislator per term. The variable ranges from 0 to 3,359. Second, we estimated the models using only the PQs issued to the national government and entities responsive to it. Table 2 shows that the share of PQs issued to the national government has declined over time, from a high of 80.6% in 2006–2010 to a low of 59.3% in 2014–2018. Table 2 shows descriptive statistics on the number of all PQs issued per legislative term and PQs issued to the national government and national government agencies per legislator.

Table 2. PQs by Legislator by Term, Chamber of Deputies of Chile, 2006–2018

Source: The authors, with data from the Chamber of Deputies.

The independent variable for Hypothesis 1 is an indicator for legislators from a ruling coalition party. The independent variable for Hypothesis 2 is a dummy variable for legislators affiliated to the president's party. The independent variable for Hypotheses 3 is an indicator for first-term legislators. The independent variable for Hypothesis 4 is political ambition. Following the standard classification, we created dummy variables for static ambition (running for re-election) and progressive ambition (running for the Senate). The omitted category are legislators who retired at the end of their terms. For the independent variable for Hypothesis 5, we use the legislator's vote share in the past election, the vote share difference between the legislator and his or her running mate, and the vote share difference between the legislator's coalition and the other major multiparty coalition. As two-seat open-list proportional representation systems made legislators run against their coalition partners, the margin of victory points to how safe the legislator – or the coalition – felt about that seat. The independent variable for Hypothesis 6 is the log value of the distance between the district and the capital city of Santiago.

As control variables, we use the district's population and the number of bills authored by each legislator – as the authoring of bills in Chile has been linked to constituency service and career advancement (Escobedo Aránguiz and Navia 2020). Following Alemán et al. (2018), we also control for committee membership in the powerful finance and constitution committees. We also included dummy variables for each legislative term and for the five largest parties in Congress. Table 3 shows the descriptive statistics for the variables.

Table 3. Descriptive Statistics for Dependent and Independent Variables

Source: The authors, with data from the Chamber of Deputies.

Notes: DV = dependent variable; PDC = Christian Democratic Party; PPD = Party For Democracy; PS = Socialist Party; RN = National Renewal; UDI = Independent Democratic Union.

Table 4 shows that the average number of PQs issued by legislators increased almost four-fold between the 2006–2010 and 2014–2018 terms. Relative to other parties in the same term, legislators from ruling coalition parties sent fewer PQs. Legislators from the PS sent more PQs than legislators from other centre-left parties. Even though in 2006–2010 and 2014–2018 the president was Michelle Bachelet, a member of the PS, legislators from that party systematically outdid other centre-left legislators in issuing PQs. In turn, legislators from the RN sent fewer PQs than UDI legislators when Sebastián Piñera (who won the election as a member of the RN) was the president. About two-thirds of PQs were sent to national government entities. Legislators from the ruling coalition or from the president's party did not behave notably differently in the rate of PQs they sent to the national government.

Table 4. Average Number of PQs by Legislator per Party, Chamber of Deputies of Chile, 2006–2018

Source: The authors, with data from the Chamber of Deputies.

Notes: Figures for the president's party is in bold, and the ruling coalition parties are in italics. PDC = Christian Democratic Party; PPD = Party For Democracy; PS = Socialist Party; RN = National Renewal; UDI = Independent Democratic Union.

Inferential analysis

Table 5 shows the estimations for generalized least squares (GLS) negative binomial models where the dependent variable is the number of all PQs issued and the number of PQs issued to the national government and national government entities. Hypothesis 1 postulates that ruling coalition legislators send fewer PQs than legislators from opposition parties. The results in all eight models are consistent with that expectation. There is no difference in the behaviour of legislators issuing PQs to the national government or to other government entities. As they issue fewer PQs, ruling coalition legislators might have other tools for constituency service and name recognition. The results of the models do not invalidate the constituency service function of PQs, but they lend support to the argument that PQs are also used – more by opposition than by ruling coalition legislators – as an oversight accountability tool to check on the executive and other government entities.

Table 5. GLS Negative Binomial Models on the Number of PQs Issued by Members of Chile's Chamber of Deputies, 2006–2018

Source: The authors, with data from the Chamber of Deputies (see methodology section).

Notes: *** significant at 0.01; ** significant at 0.05; * significant at 0.1. PDC = Christian Democratic Party; PPD = Party For Democracy; PS = Socialist Party; RN = National Renewal; UDI = Independent Democratic Union.

Hypothesis 2 suggests that legislators from the president's party send fewer PQs in general but also send fewer PQs to the national government. The results of the models do not offer conclusive evidence. In Models 1–4, which include dummies for specific political parties, the coefficient for the president's party is negative. But in the other models, without those dummies, the effect is positive, but not significant. The issuance of PQs seems to respond to coalition-level incentives rather than party-level incentives.

The next four hypotheses are related to individual-level incentives rather than to party-level incentives. The models offer no conclusive evidence in support of Hypothesis 3. Although the effect of first-termers is positive, it is not always significant. Hypothesis 4 postulates that legislators with static and progressive ambition send more PQs than legislators with discrete ambition. The results deviate from the expectations as there is no effect by the career ambition of legislators on the issuance of PQs. Our findings directly contradict those of Alemán et al. (2018). Since we rely on a longer time period and include all PQs issued – not just those publicly announced on the floor of the chamber – we are confident in qualifying their findings.

Hypothesis 5 suggests that legislators with lower vote shares in the past election send more PQs. The results are not consistent with the expectation. Legislators who might feel electorally vulnerable do not make more use of PQs. The fact that, during those three terms, the electoral system – open-list proportional representation with a district magnitude of 2 – made legislators compete against their coalition partners more than against candidates from other coalitions might point to the party affiliation and coalition affiliation of the legislator as the main determinants in the issuance of PQs. Legislators' issuance of PQs seems to respond more to accountability concerns than to constituency-service or name-recognition goals.

The effect of the intra-list and intercoalition vote margins in the past election are nuanced. A higher intra-list vote margin – a higher advantage over your list partner – has a negative impact on the issuance of PQs. Legislators who felt less threatened by their coalition partners in the last election send fewer PQs. In turn, legislators whose coalition obtained a higher vote share than the other coalition tend to issue more PQs. Legislators whose coalition is strong in a district might aspire to get both seats in the next election and thus become more active in the issuance of PQs. The fact that, in the model that includes both variables, their effect is not statistically significant points to the impacts of both variables cancelling each other out. That would signal that legislators care about their potential rivals, but their rivals are different depending on whether the legislator's seats are threatened more by a list partner or by the strength of the other coalition.

Models 5 and 6, which include the legislator's vote share and the intra-list margin of victory, show that legislators who get a higher vote share send more PQs, but when their margin of victory over their list partner is larger – when they feel safer in their seats – they send fewer PQs. Thus, there is some evidence that feeling that their seats are not as safe induces legislators to send more PQs, but the evidence is not conclusive.

Consistent with Hypothesis 6, legislators from districts further away from the capital issue more PQs than the rest. Legislators who represent centrally located districts might have easier access to government authorities and can obtain more expeditious responses to their constituencies' demands and needs than those from faraway regions. A complementary explanation is that legislators in faraway regions can disseminate their PQs in local media more easily than legislators from central districts, where the national news often overshadows local news.

Some of the control variables also offer interesting results. The district's population impacts negatively the issuance of PQs. Legislators who author more bills are more active in issuing PQs. This result is interesting, as the authorship of bills in Chile has also been found to be a constituency service tool (Alemán et al. 2018). Membership of the finance and constitution committees does not affect the issuance of PQs. In models included in the Supplementary Material, we also estimated membership in several other committees and consistently find no effect of committee membership on the issuance of PQs.

Surprisingly, legislators from the PS send more PQs, despite the fact that, in two of the three presidential terms, the president of the republic was a member of the PS. Right-wing UDI legislators send fewer PQs to the national government, though the coefficients are less strongly significant. The controls by party affiliation point to some idiosyncratic variables in how legislators from different parties' issue PQs. Though all legislators can issue PQs, not all parties use this tool equally. We included control variables for presidential terms as the overall number of PQs drastically increased in the 2014–2018 term.

The models consistently show that opposition coalition legislators send more PQs than ruling coalition legislators and that legislators who are both in the ruling coalition and the president's party send fewer PQs than the rest. The models also show that some individual traits, like being in their first term, also explain the issuance of more PQs. While some constituency service variables have an effect, the strongest determinant of the issuance of PQs in Chile was the legislator's affiliation with the government coalition.

Conclusions

Maybe because they are less common, the issuance of PQs in presidential systems has received less attention than the issuance of PQs in parliamentary systems. As a presidentialism system with an institutionalized multiparty structure in the legislature, Chile offers a unique opportunity to study the individual- and party-level incentives that account for the issuance of PQs. We analysed the impact of affiliation to the president's party and of being a member of the multiparty ruling coalition on the issuance of PQs and, specifically, on the issuance of PQs to national government entities. Legislators in the opposition send more PQs than those in the ruling coalition. Some district-level variables, like the district's distance from the capital and the district's population, also have explanatory power.

Legislators are also motivated by individual-level variables – and conduct other constituency service activities, but because coalition-level variables have strong explanatory power, we speculate that PQs might also serve as an oversight tool for the legislature to check on the actions of the executive. Future research should explore the content of PQs and assess the extent to which they serve a constituency-related service or other purposes associated with the agenda of the legislator's party or as an oversight tool for legislators to check on the executive. As PQs are a relevant tool used by legislators in some presidential systems, as in Chile, future research should explore in more detail how they are being used and what purposes they serve for ruling and opposition party legislators.